Mistletoe extract Fraxini inhibits the proliferation of liver cancer by down-regulating c-Myc expression

Mistletoe (Viscum album) is a type of parasitic plant reported to have anticancer activity including in hepatocellular carcinoma (HCC). However, the mechanism of mistletoe’s anticancer activity, and its effectiveness in treating HCC are not fully understood. We report here that mistletoe extracts, including Fraxini (grown on ash trees) and Iscador Q and M (grown on oak and maple trees), exert strong antiproliferative activity in Hep3B cells, with median inhibitory concentrations (IC50) of 0.5 µg/mL, 7.49 µg/mL, and 7.51 µg/mL, respectively. Results of Reversed Phase Proteomic Array analysis (RPPA) suggests that Fraxini substantially down-regulates c-Myc expression in Hep3B cells. Fraxini-induced growth inhibition (at a concentration of 1.25 μg/ml) was less pronounced in c-Myc knockdown Hep3B cells than in control cells. Furthermore, in the Hep3B xenograft model, Fraxini-treated (8 mg/kg body weight) mice had significantly smaller tumors (34.6 ± 11.9 mm3) than control mice (161.6 ± 79.4 mm3, p < 0.036). Similarly, c-Myc protein expression was reduced in Fraxini treated Hep3B cell xenografts compared to that of control mice. The reduction of c-Myc protein levels in vitro Hep3B cells appears to be mediated by the ubiquitin-proteasome system. Our results suggest the importance of c-Myc in Fraxini’s antiproliferative activity, which warrants further investigation.

pathway also is essential in nonalcoholic steatohepatitis-associated HCC models [14][15][16] , which suggests a central role for c-Myc in HCC, regardless of the etiology of disease. In humans, c-Myc is overexpressed in up to 70% of tumor tissues from patients with viral or alcohol-related HCC 17 , and c-Myc amplification has been linked to a more aggressive phenotype in HCC patients 18 . Sridharan and colleagues reported that c-Myc is one of four important factors that maintain the cancer stem cell phenotype in HCC 19,20 . The function of c-MYC makes it a highly attractive target for anti-cancer therapy. MYC itself is a challenging therapeutic target because of the paucity of targetable sites for the development of small molecule inhibitors thus far 21 . Small molecules have been developed to target the CMYC oncogene, however, to date these agents have not been approved clinically 22 . Collectively, these studies suggest that a pharmaceutically tractable c-Myc targeting approach would represent a novel treatment paradigm for HCC patients.
Complementary and alternative medicines are gaining more attention in oncology management 23,24 . Natural products from plants and animals were the source of medicinal preparations and, more recently, natural products have continued to enter clinical trials as anticancer and antimicrobial agents 25,26 . Natural products have been valuable sources for new therapeutic agents as 41% of FDA approved anticancer drugs are derived from natural compounds 27 . Mistletoe extract (ME; Viscum album L) can potentially spare normal cells and selectively inhibit proliferation or induce apoptosis in cancer cells. Mistletoe has been used for treating different cancers, including HCC, in Europe for over 80 years [28][29][30][31][32][33] . Different MEs are prepared from mistletoe grown on different host trees, such as fir (Abies), maple (Acer), oak (Quercus), apple. Of the various ME, Viscum Fraxini (Fraxini), grown on ash trees (Fraxinus), has the highest content of bioactive/cytotoxic mistletoe lectins (MLs) (>10 µg/ml) 32 . Intratumoral injection of Fraxini (8 mg/kg) into a human pancreatic cancer xenograft model resulted in significant tumor response with 75% of treated mice having either a partial response or complete response 34 . A phase II clinical trial reported that 20% of chemotherapy-naïve patients with advanced HCC had either complete response (13.1%) or partial response (8.1%) with limited toxicity after subcutaneous injection of Fraxini 32 . Clinical studies also have demonstrated that ME significantly reduced tumor growth and improved quality of life 35 . However, the molecular mechanisms associated with Fraxini's anticancer activity in HCC are largely unknown.
We hypothesized that the antiproliferative effect of MEs (specifically Fraxini) in HCC cells and in xenograft models is potentially through targeting the c-Myc signaling in HCC. By understanding the mechanism behind the anticancer activity of Fraxini in liver cancer, and we hope to expand our understanding and potentially identify HCC patients that would be benefit from Fraxini derived treatment.

Anti-proliferative effects of MEs in human HCC cells and xenograft model.
To identify the sensitivity of HCC cells to ME, we first examined the effects of several MEs on the Hep3B cell line. As shown in Fig. 1A, all three MEs (Fraxini, Iscador Q, and Iscador M) exerted antiproliferative activity in Hep3B cells, with median inhibitory concentrations (IC 50 ) of 0.5 µg/ml, 7.49 µg/ml, and 7.51 µg/ml, respectively. The antiproliferative effect of Fraxini was 15 times stronger than that of Iscador M and Iscador Q, respectively, in Hep3B cells. To further evaluate the efficacy of Fraxini in HCC, we conducted an antitumor efficacy study in a Hep3B cell mouse xenograft model. Fraxini treatment for 3 weeks reduced tumor volume in a dose-dependent manner. The average tumor volume in mice treated with 8 mg/kg of Fraxini (34.6 ± 11.86 mm 3 ) was 61% lower than the average tumor size of the control mice (161.6 ± 79.4 mm 3 ) (Fig. 1B,C) (P < 0.05).

Induction of apoptosis by Fraxini.
To determine the potential mechanisms by which Fraxini inhibited the proliferation of Hep3B cells, we performed cell-cycle and apoptosis analysis. Fraxini (5 µg/ml) markedly increased the number of cells in sub-G 1 /G 0 phase (500%) compared with the vehicle-treated samples ( Fig. 2A) and the apoptotic effect of Fraxini was concentration dependent. Fraxini-treated samples had more apoptotic cells than the control samples, as quantitated by Annexin V staining (Fig. 2B).
Evaluation by transmission electron microscopy of the cellular ultrastructural changes showed that Hep3B cells treated with Fraxini (5 µg/ml) had more vacuoles in the nucleus, leading to extremely condensed mitochondria in the perinuclear position of the cells (Fig. 2C, c,d), as compared to vehicle-treated cells (Fig. 2C, a,b). This finding of mitochondrial conformational changes suggests that Fraxini-treated Hep3B cells underwent apoptosis.
Fraxini modulates the apoptotic modulating protein and cell signaling protein, especially c-Myc. To investigate the effect of Fraxini on apoptotic proteins and other signaling proteins, we performed RPPA analysis in Fraxini-treated Hep3B cells (Fig. S2A). As shown in Fig. 3A, the protein levels of Bcl-xl, Bcl2, pRb and CDK1 were reduced by Fraxini, whereas cleaved caspase 7 were elevated in a concentration-dependent manner, according to RPPA analysis results. Western blot results confirmed that, of these proteins, Fraxini increased p21 and reduced Bcl-xl levels (Fig. S2B).
Fraxini modulated cell-signaling proteins, especially c-Myc, in Hep3B cells. In addition to inhibiting anti-apoptotic proteins and increasing pro-apoptotic protein expression (Fig. 3A), Fraxini (5 µg/ml) also inhibited PI3K/mTOR pathway-associated proteins such as PI3Kα, p-4EPB1 (S65 and T37), and Fox; increased GSK3β; and reduced phosphorylated NFκB and NDRG1 in a concentration-dependent manner (Fig. 3B). Furthermore, RPPA analysis showed that Fraxini treatment led to remarkable suppression of c-Myc in Hep3B cells: Fraxini (5 µg/ml) down-regulated c-Myc expression by 84%, and the reduction of c-Myc was dose-dependent (Fig. 3B). Western blot results further confirmed that Fraxini notably inhibited the c-Myc expression in a concentration dependent manner in Hep3B cells, while it also decreased expression of pRb, and p4EBP in a dose-dependent manner, which further confirmed the RPPA results (Fig. 3C). Intriguingly, Fraxini treatment also reduced c-Myc expression in Hep3B tumor tissues compared to that of control mice (Fig. 3D).

Fraxini's posttranslational down-regulation of c-Myc expression. We investigated whether Fraxini
altered CMYC gene expression to reduce c-Myc protein level in Hep3B cells. Surprisingly, CMYC gene expression was not altered by Fraxini treatment (Fig. 5A), suggesting that the effect of Fraxini on c-Myc is mediated at the translational level rather than the transcriptional level.
Studies have shown that c-Myc is degraded through the proteasomal ubiquitination with a very short half-life. Our cycloheximide chase assay results showed that the half-life of c-Myc in Fraxini-treated Hep3B cells was lower than in vehicle-treated cells (Fig. 5B). Pretreating the Hep3B cells with MG-132, an inhibitor to reduce 26S proteasome activity for 4 hr prior to treating the cells with Fraxini, we found that Fraxini failed to reduce c-Myc levels compared with cells treated with Fraxini alone (Fig. 5C).
Because a previous study showed that the phosphorylation of c-Myc at T58 and S62 plays a pivotal role in c-Myc activity, stability, and degradation 36 , we investigated the effect of Fraxini on the phosphorylation of c-Myc. As shown in Fig. 5D, Fraxini decreased phosphorylation of c-Myc at S62 and slightly increased phosphorylation of c-Myc at T58. To confirm that Fraxini's anticancer activity was mediated by posttranslational regulation of c-Myc, especially via regulating c-Myc phosphorylation, we tested the antiproliferative effect of Fraxini in Burkitt lymphoma Raji cells, which harbor CMYC T58 mutation, resulting in c-Myc stabilization 37 . Strikingly, Fraxini (up to 20 µg/ml) exerted minimal antiproliferative activity in Raji cells (Fig. 5E), which correlates with the lack of down-regulation of c-Myc expression (Fig. 5F). www.nature.com/scientificreports www.nature.com/scientificreports/

MLs and Fraxini-elicited anti-proliferative activity and down-regulation of c-Myc expression.
To identify potential compounds responsible for Fraxini-elicited anticancer activity in HCC, we investigated the effect of water-soluble and lipid-soluble fractions of Fraxini on the growth of Hep3B cells. Proliferation of Hep3B cells was inhibited by the water-soluble fraction of Fraxini, which was similar to the anti-proliferative effects of Fraxini, but the lipid-soluble fraction of Fraxini showed minimum anti-proliferative activity in these cells (Fig. 6A). The water-soluble fraction of Fraxini also induced down-regulation of c-Myc protein expression (Fig. 6B). Further fractionation of the water-soluble components of Fraxini revealed that fraction 7 was enriched in mistletoe lectins (MLs) analyzed by the proteomic core at MDACC (Tab. S1), and was the most effective at inhibiting the proliferation of Hep3B cells with IC 50 ≤ 1 ng/ml ML compared with the other fractions (Supporting Information Fig. S3). This finding suggests that MLs could be the bioactive components responsible for Fraxini's anticancer activity in HCC cells.
Because MLs have anticancer and immunomodulation effects 38,39 , we treated Hep3B and PLC cells with MLs and observed that MLs reduced cell growth in both cell types in a dose-dependent manner. Intriguingly, MLs exerted much stronger anti-proliferative activity in Hep3B cells (IC 50 < 1 ng/ml) than in PLC cells; a 6 times higher concentration of ML was required to achieve similar activity in PLC cells (Fig. 6C). Similarly, ML treated Hep3B cells underwent apoptotic cell death in a concentration dependent manner (Fig. 6D). Strikingly, a ML concentration as low as 1 pg/mL dramatically decreased c-Myc expression in Hep3B cells, and this down-regulation of c-Myc expression was dose-dependent (Fig. 6E). Additionally, we also observed that proteasome inhibition prevented reduction of c-Myc protein induced by 1 pg/ml ML (Fig. 6F). Together, all these data augment that MLs are indeed responsible for Fraxini's antitumor activity in HCC cells.

Discussion
We demonstrated that MEs had antiproliferative activity in HCC cell lines; of all the MEs studied, Fraxini has the most potent antiproliferative activity. Fraxini's antiproliferative activity in Hep3B cells was associated with inducing apoptosis and mitochondrial conformational changes. Our xenograft mouse model results revealed that Fraxini suppressed Hep3B tumor development in vivo. By analyzing the proteomic profile of Fraxini-treated Hep3B cells, we found that Fraxini had profoundly inhibitory effect on c-Myc protein expression. We also found that MLs exerted stronger antiproliferative activity and down-regulated c-Myc expression in Hep3B cells. Given that c-Myc is crucial for HCC development and progression regardless of the etiology of disease and that MLs already have been tested in clinical trials for various tumors, our findings suggest that Fraxini could be a potentially efficacious treatment for patients with advanced HCC or with overexpression of CMYC gene. Further studies are needed to determine the effects of Fraxini in patients with HCC. To our knowledge, we are the first to identify the Fraxini's anticancer activity could potentially be mediated through down-regulation of c-Myc pathway. www.nature.com/scientificreports www.nature.com/scientificreports/ Despite long-term use of ME in cancer treatment, the mechanism of ME's anticancer activity remains elusive. Previous studies have reported some mechanisms of the anticancer activities of MEs that primarily focus on ME's role in immunomodulation. For example, several studies have shown that aqueous MEs can activate monocytes/ macrophages, granulocytes, natural killer cells, and helper T cells and can induce various cytokines [40][41][42][43][44] . Viscum album L extract exerts cytotoxic and growth-inhibiting effects on various human tumor cell lines, lymphocytes, and fibroblasts in vitro by inducing apoptosis or necrosis 45,46 . Similarly, we also found that Fraxini suppressed the proliferation of Hep3B cells by inducing apoptosis and mitochondrial conformational changes. As we have also observed more vacuoles in the cell nucleus with Fraxini treatment compared to that of the vehicle treated cells, which may suggest Fraxini could also inhibit the proliferation through induction of autophagic cell death. However the examination of the expression of LC3 protein, which is a hallmark of autophagic cell death did not show increase of LC3 protein in Hep3B cells after being treated with Fraxini (Data not shown). The inhibitory effect of Fraxini on cell growth in Hep3B cells was primarily mediated through induction of apoptosis in human HCC cells. Further mechanistic study using proteomic analysis demonstrated that Fraxini could potentially affect a number of cancer cell survival and signaling pathways as illustrated in Figs 3A,B and 7 with c-Myc down-regulation being the most effective in Hep3B cells, which has not been reported previously.
CMYC is one of the most commonly activated oncogenes in the development of human liver cancer 47 and is involved in diverse biologic processes, such as cell growth, apoptosis, and metabolism. Tumorigenic transcriptional factors such as KRAS, AKT, PTEN, and p53, which play key roles in cell proliferation and apoptosis, are regulated by c-Myc 48 . In mice, overexpression of CMYC can induce HCC, whereas inhibition of CMYC expression results in regression of liver tumors through the dedifferentiation of mature hepatocytes that give rise to tumors 49 . Given the potent effect of c-Myc on tumor development, it is not surprising that the cellular expression of c-Myc is regulated transcriptionally (initiation and elongation), post-transcriptionally (mRNA stability and translation), and posttranslationally (protein stability) 50 . Furthermore, in the N-terminus of c-Myc, two phosphorylation sites (T58 and S62) are critical in c-Myc protein stability. Sears et al. reported the phosphorylation mechanisms of T58 and S62 by Ras-activated signaling pathways ERK and GSK3β 51 . Our data indicate that Fraxini reduced c-Myc protein but not mRNA expression and slowed Hep3B tumor growth in mice. It appears that Fraxini suppressed c-Myc by decreasing c-Myc stability, which may involve the phosphorylation of both T58 and S62 on c-Myc (Fig. 7). We speculate that Fraxini regulates c-Myc stability in Hep3B cells; however, the exact mechanism by which Fraxini reduces c-Myc activity needs to be further investigated.
ML, especially MLI, has been recognized as a major bioactive component responsible for ME's anticancer activity in various types of cancer cells 45,52,53 . In fact, several studies have shown that Aviscumine, a human www.nature.com/scientificreports www.nature.com/scientificreports/ recombinant MLI, induces significant cytotoxicity in various types of cancer cells by inducing either apoptosis or necrosis [54][55][56] . One study showed that Aviscumine activated the MAP kinase pathway, p38, Erk1/2, and SAP/JNK in ovarian cancer SKOV3 cells; activated caspases and two MAP kinase pathways in keratinocytes in vitro; and inhibited protein synthesis by inactivating the 60 S subunit of ribosomes 39 . In our study, ML strongly inhibited the proliferation of Hep3B cells (which had the highest c-Myc expression of the cell lines studied) but had moderate effect on PLC cells (which had the lowest c-Myc levels). These results were similar to those of Fraxini-treated cells, suggesting that the antitumor activity of both Fraxini and MLs are very likely mediated by c-Myc. Indeed, we also observed that MLs at a very low concentration (1 pg/ml) notably down-regulated c-Myc expression in Hep3B cells, which further supports the hypothesis that MLs are the bioactive components responsible for Fraxini's antiproliferative and proapoptotic activity. Whether other bioactive components in Fraxini could contribute to Fraxini's anticancer activity remains to be determined.
Mistletoe extracts has been studied in various clinical trials in different counties. However, across these studies, not only did the mistletoe extracts have different pharmaceutical manufacturing processes and formulations, but also the route of delivery and doses varied considerably 57 . In these reports, mistletoe extracts were administered subcutaneously, intravenously, or intravenously and subcutaneously combined. A Meta-analysis found a positive treatment effect, but did report several methodological limitations regarding the retrospective study design 58 . In terms of adverse effect, the subcutaneous administration showed more side effects, while oral administration of mistletoe preparations with apparent lack of adverse reactions might be a promising adjuvant alternative therapy for some cancer patients 59 . Although clinical studies of mistletoe are limited, a phase II study with HCC patients reported that about 20% of chemo-naïve advanced HCC patients had either CR or PR with limited www.nature.com/scientificreports www.nature.com/scientificreports/ toxicity after subcutaneous injection of Fraxini 13,60 . Our data here further supports a possible clinical benefit of Fraxini treatment in HCC patients. Therefore the efficacy of Fraxini and possibly the human recombinant ML, such as Aviscumine that has proper quality control minimizing batch to batch variation, in HCC patients warrants further investigation.
We showed for the first time that Fraxini potently inhibits the proliferation of HCC cells and of tumor tissue in a xenograft mouse model primarily by destabilizing c-Myc. Further research is necessary to determine whether patients with CMYC-overexpressing HCC show better responses to Fraxini treatment than other HCC patients without overexpression of CMYC.
Cell culture. Hep3B, Hep G2, and PLC/PRF/5 (PLC) cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA) and were maintained in a humidified atmosphere containing 5% carbon dioxide at 37 °C. Hep3B and HepG2 cells were routinely cultured in Dulbecco modified Eagle medium/F12. PLC cells were cultured in minimum essential medium. All media were supplemented with 10% heat-inactivated fetal bovine serum (HyClone Laboratories, Inc., Logan, UT), penicillin (10 IU/ml, Invitrogen/Thermo Fisher Scientific, Carlsbad, CA), and streptomycin (10 µg/ml, Invitrogen/Thermo Fisher Scientific). Human Burkitts B-cell lymphoma Raji cells were also purchased from ATCC (Manassas, VA) and were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS; Hyclone, Logan, UT), 1 mM L-glutamine, and 50 IU/ml penicillin and 50 μg/ml streptomycin. Mouse primary HCC cells (EC4) was a gift from Dr. Dean W. Felsher at Stanford University. The cells were maintain in DMEM culture medium supplemented with 10% heat-inactivated fetal bovine serum (HyClone Laboratories, Inc., Logan, UT), L-glutamine (2 mM, Gibco, Grand Island, New   Center Animal Care and Use Committee. Six-week-old female BALB/c athymic (Nu/Nu) mice were purchased from the Department of Experimental Radiation Oncology at MD Anderson Cancer Center and were acclimated to the institutional animal care facility for 1 week. Mice were injected subcutaneously with Hep3B cells (5 × 10 6 per mouse) and were placed in one of three treatment groups (n = 5 for each group): control or Fraxini (4 mg/kg or 8 mg/kg). Each group was treated via subcutaneous injection twice per week starting the day after tumor cell inoculation. Tumor take rate was 100%. Tumor growth was then measured for 3 weeks. The length and width of the tumors (mm) were measured twice per week using calipers. The tumor volume was calculated using the formula (l × w 2 ) × 0.5, where l and w represent the length and the width, respectively.
Cell-cycle and apoptosis analysis. For cell-cycle analysis, Hep3B cells (2.5 × 10 6 ) grown in 100-mm dishes were treated with Fraxini (1 µg/ml-20 µg/ml) for 24 hr. Cells were subjected to trypsinization and centrifugation, and the pellets were suspended and washed in 1X phosphate-buffered saline solution (PBS) and were fixed overnight in 70% ethanol at 4 °C. The cells were then washed with 1X PBS and were suspended in PBTB staining solution containing PBS, 0.5% bovine serum albumin, 0.005% Tween-20, propidium iodide (10 μg/ml), and DNase-free RNase (1 μg/ml). Cells were incubated in the dark for 30 min at 37 °C before fluorescence-activated cell-sorting analysis (FACS) using a BD FACSCaliber flow cytometer (BD Biosciences, San Jose, CA). The percentage of cells in each phase of the cell cycle was estimated from the DNA histogram content.
The apoptotic cell death was further measured by Annexin V surface staining in Hep3B cells treated with Fraxini for 48 hr. Briefly, cells (2.5 × 10 6 ) were double-stained with fluorescein isothiocyanate-conjugated Annexin V and propidium iodide per the manufacturer's instructions (BD Biosciences). Fluorescence was detected by a BD FACSCalibur flow cytometer and was analyzed using CellQuest TM Pro software (BD Biosciences). transmission electron microscopy. We investigated whether Fraxini could alter intracellular structures in Hep3B cells. Cellular structure was examined by transmission electron microscopy (TEM). Hep3B cells were grown on glass coverslips for 24 hr before treatment with Fraxini (10 µg/ml). After incubating for 24 hr, cells were then washed with cold PBS and were prepared for electron microscopy, as previously described 63 .
Reverse phase protein array. Hep3B cells (1 × 10 6 ) were treated with Fraxini (5 µg/ml, 10 µg/ml, or 20 µg/ml) for 24 hr and were harvested, and cell lysates were subjected to reverse phase protein array (RPPA) analysis by the Functional Proteomics Core Facility at The University of Texas MD Anderson Cancer Center. Briefly, cells were washed twice with PBS and were suspended in ice-cold lysis buffer (1% Triton X-100, HEPES [ , and 10% glycerol) supplemented with proteinase inhibitors (Roche Applied Science, Madison, WI), were diluted and subject to proteomic array analysis, as previously described 64,65 . Quantitative reverse transcription-polymerase chain reaction. We used quantitative reverse transcription-polymerase chain reaction (RT-PCR) analysis to compare CMYC gene expression in Hep3B cells. Total RNA was prepared from Hep3B cells treated with Fraxini with the doses indicated in the figure for 1 day, and cDNA was generated. RT-PCR primers for MYCC were as previously described 66 : Forward: 5′-TACCCTCTCAACGACAGCAG-3′ Reverse: 5′-TCTTGACATTCTCCTCGGTG-3′ Primers for β-actin were used 67 : Forward: 5′-GATGATGATATCGCCGCGCTCGTCGTC-3′ Reverse: 5′-GTGCCTCAGGGCAGCGGACCGCTCA-3′ RT-PCR was done following the manufacturer's protocol (Invitrogen/Thermo Fisher Scientific). Target gene expression levels were normalized to the β-actin expression level for each sample. PCR conditions for CMYC included an initial denaturation step (4 min at 95 °C) followed by 35 cycles of denaturation (94 °C for 30 sec), annealing (60 °C for 30 sec), annealing (72 °C for 45 sec), and a final extension step (72 °C for 5 min). The conditions for β-actin were the same.
Immunoblotting. Hep3B cells were treated with Fraxini, water-soluble fraction of Fraxini (preparation procedure of water-soluble fraction of Fraxini was illustrated in Fig. S1), or MLs for 24 hr and then were harvested in sodium dodecyl sulfate buffer. Total protein (50 μg per well) was separated on a 10-15% sodium dodecyl sulfate gel and was transferred onto polyvinylidene fluoride membranes. The membranes were blocked with 5% nonfat milk in Tris-buffered saline solution and then were probed with primary antibodies at 4 °C overnight. The membranes were visualized using SuperSignal substrate (Pierce ECL Plus, Thermo Fisher Scientific). β-actin was detected for normalization of results. siRNA transfection. Hep3B cells were plated in 6-well plates and allowed to attach overnight. Transient transfection of nonspecific siRNA (control siRNA) and c-Myc siRNA (OriGene, Rockville, MD) was carried out using Lipofectamine RNAiMAX Transfection Reagent (Thermo Fisher Scientific) following the manufacturer's instructions. Beginning 48 hr after transfection, cells were treated with Fraxini for 24 hr. Protein was collected for Western blot analysis.